™Daily Brief · Nov 12
2026.11.12 Thursday
14 AI stories worth your time, each with one line on why it matters.
ListenNov 12 · 14 stories
In this issue
Nov 12 · 14 stories-
Fed's Bowman Warns Weak AI Returns Could Trigger a Sharp Market Correction
Federal Reserve Vice Chair for Supervision Michelle Bowman said on Nov 10 that equity markets could face a sharp correction if AI investment returns fall short of expectations, with AI-related debt on track to exceed $8…
Why it mattersThe Fed's top supervision official put AI debt and circular financing on the list of main economic risks, citing $800 billion in debt and off-balance-sheet exposure of more than t…
-
Google Launches Gemini Decision 1: A Model That Answers Instead of Writing
Google launched the Gemini Decisions API on Nov 11; Gemini Decision 1 answers 1,000 questions in one pass with confidence scores, at 150 to 250 milliseconds.
Why it mattersGoogle has split the judgment agents make most often out of generative calls, pricing a single decision in milliseconds and fractions of a dollar for the first time.
-
United Airlines Bans AI Agents From Booking Flights
United Airlines said on Nov 11 it will bar unauthorized third-party AI agents from booking, changing or checking in flights for passengers, with offending reservations subject to cancellation.
Why it mattersUnited is the first U.S.
-
ByteDance Readies Muse-Like AI Assistant; Zhang Sees Frontier Catch-Up by 2029 at Earliest
ByteDance plans to launch a personal AI assistant similar to Meta's Muse in the coming weeks, Bloomberg reported, and co-founder Zhang Yiming told staff that catching up with OpenAI and Anthropic at the frontier would t…
Why it mattersByteDance is adding a personal-assistant layer on top of Doubao's roughly 324 million monthly active users, while its founder set the frontier catch-up date at 2029 or 2030.
-
Baidu Launches ERNIE 5.5 Flash Preview, Weights Open by Nov 25
Baidu released ERNIE 5.5 Flash Preview on Nov 11, a 280-billion-parameter model activating 14 billion per token with a 1-million-token context; the weights will be open-sourced under Apache 2.0 before Nov 25.
Why it mattersSix days after open-sourcing the base weights on Nov 5, Baidu shipped the first post-trained version, with the weights promised under Apache 2.0 before Nov 25.
-
Anthropic Says AI-Driven R&D Acceleration Has Met Its Safety-Policy Threshold
Anthropic said in its amended IPO filing dated Nov 10 that it has crossed the "dramatically accelerated progress in AI research and development" threshold in its Responsible Scaling Policy, based on expectations for fut…
Why it mattersA week before listing, Anthropic formally recorded in its prospectus that its own safety policy's R&D-acceleration threshold has been met, while the safeguards that follow have no…
Follow HIMIAO Horizon
No sign-up · Unsubscribe any timeNotifications
A ping on your phone or computer when the brief is out. No email needed.
